Modaltune: Fine-Tuning Slide-Level Foundation Models with Multi-Modal Information for Multi-Task Learning in Digital Pathology

Prediction tasks in digital pathology are challenging due to the massive size of whole-slide images (WSIs) and the weak nature of training signals. Advances in computing, data availability, and self-supervised learning (SSL) have paved the way for slide-level foundation models (SLFMs) that can improve prediction tasks in low-data regimes. However, current methods under-utilize shared information between tasks and modalities. To overcome this challenge, we propose ModalTune, a novel fine-tuning framework which introduces the Modal Adapter to integrate new modalities without modifying SLFM weights. Additionally, we use large-language models (LLMs) to encode labels as text, capturing semantic relationships across multiple tasks and cancer types in a single training recipe. ModalTune achieves state-of-the-art (SOTA) results against both unimodal and multi-modal models across four cancer types, jointly improving survival and cancer subtype prediction while remaining competitive in pan-cancer settings. Additionally, we show ModalTune is generalizable to two out-ofdistribution (OOD) datasets. To our knowledge, this is the first unified fine-tuning framework for multi-modal, multitask, and pan-cancer modeling in digital pathology. Our code is publicly available at https://github.com/ martellab-sri/ModalTune.